Manganiferous Sediments, Rocks, and Ores

Manganiferous Sediments, Rocks, and Ores

7.11 Manganiferous Sediments, Rocks, and Ores J. B. Maynard University of Cincinnati, OH, USA 7.11.1 7.11.2 7.11.3 7.11.4 7.11.5 CHEMICAL FUNDAMENTAL...

379KB Sizes 0 Downloads 11 Views

7.11 Manganiferous Sediments, Rocks, and Ores J. B. Maynard University of Cincinnati, OH, USA 7.11.1 7.11.2 7.11.3 7.11.4 7.11.5

CHEMICAL FUNDAMENTALS DISTRIBUTION OF MANGANESE IN ROCKS AND NATURAL WATERS COMPOSITION OF MANGANESE ACCUMULATIONS COMMON MANGANESE MINERALS BEHAVIOR OF MANGANESE IN IGNEOUS SETTINGS, ESPECIALLY MID-OCEAN RIDGE VENTS 7.11.6 BEHAVIOR OF MANGANESE IN SEDIMENTATION 7.11.6.1 Manganese Nodules and Crusts in Modern Sediments 7.11.6.2 Manganese Carbonates in Modern Sediments 7.11.7 A GENERAL MODEL OF SEDIMENTARY MANGANESE MINERALIZATION 7.11.8 BEHAVIOR OF MANGANESE IN SOILS AND WEATHERING 7.11.9 CONCLUSIONS REFERENCES

Manganese is the ninth or tenth most abundant element in the Earth’s crust (depending on the model used for crustal composition). Most of its industrial use is in steel making with a much lesser amount going into the production of batteries. It is very similar to iron in its chemical properties. Both are commonly found with þ 2 and þ 3 valences with high spin states for the 3d electrons and with similar ionic radii. Mn and Fe þ 2 ions ˚ and 0.78 A ˚ , while the þ 3 ions have radii 0.83 A ˚ and 0.65 A ˚ , respectively (Li, 2000, have 0.70 A tables 1 –4). Accordingly, manganese is commonly found substituted in small amounts in iron minerals. Manganese, however, also has access to a higher valence state, þ 4, which gives rise to a plethora of complex manganese oxide minerals. Both elements are mined from supergene-enriched sedimentary deposits of a great variety of ages. Iron, however, is dominantly produced from Archean to very early Paleoproterozoic deposits, whereas manganese ores come from post-Archean rocks.

289 293 294 297 298 300 300 301 302 303 305 305

7.11.1 CHEMICAL FUNDAMENTALS As with iron, manganese geochemistry in sedimentary environments is governed by oxidation and reduction. Eh – pH relations show a relatively large field of stability for dissolved Mn2þ compared to the solid oxides (Figure 1). At the pH value of 8 for seawater, or of 5– 7 for fresh surface waters, manganese should be soluble except under strongly oxidizing conditions. The addition of carbonate species to the system creates a large region in which solid manganese, in the form of rhodochrosite, is stable under reducing conditions (Figure 2). Unlike FeS or FeS2, the sulfides of manganese (alabandite and hauerite) are very soluble under reducing conditions, so manganese behavior at low Eh is controlled by carbonate minerals, in contrast to iron, which is controlled more by the sulfides. In freshwater, the pH is normally too low for rhodochrosite precipitation, but for seawater, a slight increase in the amount of CO22 should lead to rhodochrosite 3

289

290

Manganiferous Sediments, Rocks, and Ores

Figure 1 Eh – pH diagram showing the relationships among Mn oxides. Note the large field of stability for dissolved Mn2þ and low Eh and pH.

formation if there is a supply of Mn. MnCO3 minerals have, in fact, been reported from a variety of low Eh marine environments. As first pointed out by Krauskopf (1957), Eh– pH diagrams suggest a mechanism for separating manganese from iron. Figure 3 shows that soluble manganese has a considerably larger stability field than soluble iron under moderately reducing conditions. Because almost all sediments become at least slightly reducing a few centimeters below the sediment – water interface, Mn2þ is commonly mobilized into the pore water, while iron remains fixed as an oxide or hydroxide. Under conditions of low Eh and high sulfur content, such as in reducing marine sediments, the iron is fixed as a sulfide, but manganese is still mobile. Dissolved Mn2þ diffuses upward, and is either precipitated at the sediment – water interface by the oxygen in the bottom water, or dispersed into the overlying water and carried to other parts of the basin, depending on the oxygen content of the water. In soils, the greater mobility of manganese will lead to a greater tendency to move downwards in profiles, leaving behind an iron-rich crust. If the

manganese is not lost from the soil altogether, its downward movement can produce significant enrichments (e.g., Sivaprakash, 1980; Roy, 1981, pp. 124 – 132). The great majority of mineable manganese ores have undergone supergene enrichment (Varentsov, 1996). The variety of manganese phases found in sediments and the common presence of metastable phases suggests that kinetic factors are important in manganese geochemistry. Unlike most other metals, considerable work has been done on the geochemical kinetics of manganese (Hem, 1972, 1981; Glasby, 1974; Stumm and Morgan, 1996). For most manganese oxides, precipitation involves oxidation of the Mn2þ in solution to Mn4þ. Although thermodynamically favored, this transition is strongly inhibited kinetically. Oxidation of Fe2þ can be modeled by the expression 2d½Fe2þ =dt ¼ k½Fe2þ ½O2 ½OH2 2

ð1Þ

which involves only solution species. Mn2þ oxidation is a heterogeneous reaction that follows

Chemical Fundamentals

291

Figure 2 Eh – pH diagram for Mn carbonate and oxides. Rhodochrosite should be the dominant mineral at high pH under low Eh conditions.

an autocatalytic relationship 2d½Mn2þ =dt ¼ k0 ½Mn2þ  þ k1 ½Mn2þ   ½MnO2 ½O2 ½OH2 2

ð2Þ

This expression tells us that the precipitation of manganese oxides is favored by the presence of a pre-existing surface of manganese oxide or, as it turns out, of iron oxide. These oxide surfaces have the ability to sorb appreciable quantities of ions from solution, particularly favoring the cations of the transition metals. Oxidation of Mn2þ involves, first, a sorption of the ion onto the oxide surface, followed by the oxidation step, hence the importance of a pre-existing surface. Manganese and iron oxidation also differ in their pH dependence: iron oxide forms at an appreciable rate at pH values above 6, whereas the equivalent rate for manganese oxide is not reached until pH ¼ 8.5 (Stumm and Morgan, 1996, figure 11.6). Thus, in seawater (pH ¼ 8), there will be a tendency for iron to precipitate, but for manganese to remain in solution, even when thermodynamic considerations suggest that both should precipitate. The catalytic effect of oxide surfaces is therefore especially important in seawater. In fact, most manganese nodules appear to have nucleated

around a grain such as a shark’s tooth, and these nuclei commonly have a rim of iron oxide or hydroxide that precedes the deposition of the manganese oxide layers (Burns and Brown, 1972). The constants in Equations (1) and (2) were determined in batch experiments and are difficult to apply to natural systems with flowing water, but Hem (1981) has calculated that, at a pH of 8.5, it would take nearly a million years to form an oxide layer 0.1 mm thick. Note that these kinetic constraints do not apply to manganese carbonates, which form directly from Mn2þ. Balzer (1982) has shown that Mn2þ concentrations in bottom waters of the Baltic rise above the level predicted by equilibrium with oxides as the underlying sediment becomes anoxic, and stabilize at about the level predicted for MnCO3 saturation. The high capacity of manganese oxides for the sorption of cations leads to an enrichment of manganese-rich sediments in a number of economically valuable transition metals, particularly copper, nickel, and cobalt. However, the exact mechanism of the incorporation of these metals remains controversial. For example, R. G. Burns and M. Burns (1977a) maintained that these elements substitute within the lattice of the manganese minerals, whereas Glasby (1974)

292

Manganiferous Sediments, Rocks, and Ores 0.8 NO3– /N2 0.6 MnO2 /Mn2+

Volts

0.4 0.2 0 –0.2

Fe(OH)3 /Fe2+ SO42– /H2S

–0.4

Figure 4 Sequence of redox reactions experienced in modern sediments.

appear to exhibit catalysis, including Bacillus and Leptothrix, and a variety of manganese oxide minerals can be produced. Zhang et al. (2002) have shown that the oxidation process follows standard Michaelis– Menten kinetics: 2 Figure 3 Eh – pH diagram with sulfides included. Note the large field of stability for Fe sulfide compared to Mn sulfide. The points labeled shallow and deep refer to typical compositions of waters from the Black Sea (reproduced by permission of Society of Economic Geologists from Sedimentary and Diagenetic Mineral Deposits: A Basin Analysis Approach to Extraporation (eds. E. R. Force, J. J. Eidel, and J. B. Maynard), 1991, pp. 147– 159).

argued that this cannot be the case and proposed instead that the metals are loosely held on exchange sites. The influence of microorganisms is another problem in the geochemistry of manganese. Bacteria catalyze the oxidation of Fe2þ (Nealson, 1997), particularly at low pH where the abiotic reaction is slow. The slow abiotic precipitation rates of manganese oxides also provide an opportunity for bacterial mediation of oxidation (Cowen and Bruland, 1985), and a number of workers have reported bacterial catalysis of manganese oxidation. For example, Mandernack and Tebo (1993) showed that manganese removal from Galapagos vent fluids was strongly inhibited by sodium azide, an agent toxic to bacteria. Juan de Fuca vent fluids did not show this effect, suggesting that the proportion of manganese removed by bacteria catalysis varies greatly between sites. Manganese removal can be produced both by direct oxidation of Mn2þ (Mandernack et al., 1995) and by sorption onto the surface of bacteria (Roitz et al., 2002). Various bacterial groups

d½Mn2þ  Xðk½Mn2þ Þ ¼ dt ðKs þ ½Mn2þ Þ

ð3Þ

where X is the cell concentration in mg L21, k is the maximum Mn2þ oxidation rate in micromoles of Mn2þ (mg cells min)21, and Ks is the so-called “half-velocity constant,” which is the concentration of Mn2þ when the oxidation rate is half of the maximum. They calculated values for k of 0.0059 and for Ks of 5.7 at pH ¼ 7.5, dissolved oxygen ¼ 8.05 mg L21, and 25 8C. They also reported that for the abiotic reaction, k0 is ,1026 min21, so that even a modest cell density would provide faster oxidation than the homogeneous abiotic reaction. Bacteria also seem to be involved in the oxidation of cobalt (Tebo and Nealson, 1984; Moffett and Ho, 1996), which may provide a mechanism whereby these two elements become associated in manganese nodules. Bacteria also catalyze the reduction of Mn4þ to Mn2þ. Manganese reduction follows a series of reactions by which bacteria choose a succession of oxidants to metabolize organic matter. This succession produces a stratification of chemical species in sediments (Froelich et al., 1979) or a lateral zonation in aquifers (Chapelle, 1993; Vrobelsky and Chapelle, 1994). This behavior has been referred to as a redox ladder (Langmuir, 1997) and is illustrated in Figure 4. An important characteristic of these successions is that they show little overlap: manganese reduction does not occur in the presence of abundant nitrate (Dollhopf et al., 2000) and iron reduction does not begin until all of the available MnO2 has been reduced (Myers and Nealson, 1988; Lovley and Phillips, 1988; Wijsman et al., 2002). The kinetics of the process have been investigated by Dollhopf

Distribution of Manganese in Rocks and Natural Waters et al. (2000) using various strains of Shewanella putrefaciens. The appearance of Mn2þ as a bacterial product initially follows the expression ln ½Mn2þ  ¼ ln A þ BðtÞ

ð4Þ

where A and B are constants and t is the time in minutes. B was found to be independent of the strain used and to average 0.040 min21. A, which reflects the lag time before cell activity begins, varied substantially with the strain used. The results show that bacterial manganese reduction is geologically quite rapid and that the presence of organic matter in sediments or in aquifers should lead to the release of soluble manganese into the overlying water or into the down-flow portion of the aquifer.

7.11.2 DISTRIBUTION OF MANGANESE IN ROCKS AND NATURAL WATERS An examination of the distribution of manganese among the various reservoirs that make up the Earth reveals much about how the element behaves in geochemical cycles. Table 1 compares manganese and iron in some common rock reservoirs and in some key rock types and types of natural waters. As shown in the previous section, the geochemistry of manganese closely resembles that of iron. Therefore, an understanding of manganese behavior, especially when it comes to the formation of ore deposits, entails an understanding of how manganese and iron differ. Table 1 Distribution of Mn and Fe. Rock type Rock reservoirs Carbonaceous chondrites Upper mantle Oceanic crust Island-arc andesites Upper continental crust Archean upper crust

Mn (ppm)

Mn/Fe

1,900

182,000

0.010

1,000 1,300 1,100 600

64,000 81,000 58,000 35,000

0.016 0.016 0.019 0.017

1,400

62,000

0.023

Rock types and natural waters Basalt 1,550 Granite 390 Shale 730 Black shale (SDO1) 325 Sandstone 850 Limestone 420 Oceanic sediment 2,700 River water 0.0082 Seawater 72 £ 1026 Source: Li (2000).

Fe (ppm)

83,000 21,100 50,000 65,300 35,000 9,500 36,000 0.040 250 £ 1026

0.019 0.018 0.015 0.005 0.024 0.044 0.075 0.21 0.35

293

The rock reservoirs on the modern Earth show a very narrow range of Mn/Fe ratios, ranging only from 0.016 to 0.019, which demonstrates how similar the two elements are in the normal terrestrial rock cycle. Carbonaceous chondrites, which are meteorites that presumably represent the primordial composition of the Earth as a whole, are enriched in iron relative to manganese compared to the Earth’s crust and mantle. This difference reflects the concentration of metallic iron, but not of manganese, in the Earth’s core. Another variation in composition from the normal crustal value of 0.017 is seen in Archean crust, which averages 0.023, a value that is higher than any common igneous rocks. When looked at by rock type (Table 1), the Mn/Fe ratio in average basalt, granite, and shale is very close to that of the crustal reservoirs of the modern Earth. Other sedimentary lithologies, however, show pronounced enrichments or depletions in manganese relative to iron, telling us that it is in the Earth’s exogenic cycles that we should look for processes that form large accumulations of manganese. In particular, note the large enrichment of manganese in limestones and the strong depletion in black shales. The cause of both is the great insolubility of iron when present in the sulfide pyrite compared to the very limited field of stability of alabandite. This strong contrast between iron and manganese behavior under anoxic conditions is well illustrated by the modern deep versus shallow water of the Black Sea (Figure 5). Dissolved iron is vanishingly low in both the shallow and deep portions of the basins. Manganese, although similarly absent from the shallow water, is much higher than iron in the deep water. There is also a peak in manganese just beneath the redox interface that reflects the redissolution of MnO2 particles that formed in the shallow water and sank through the interface. The consequence of this contrast in behavior is that iron tends to be concentrated in deeper-water shales deposited under lower oxygen levels, whereas manganese is concentrated in shallower, more oxygenated environments dominated by limestones. In fact, most large manganese ore deposits are (or were) originally manganese carbonates rather than oxides or sulfides. An ideal factory for the creation of giant manganese accumulations would be a silled basin with low-oxygen bottom waters in the tropics surrounded by fringing carbonate reefs. The giant Molango manganese carbonate deposit, in Mexico, formed as a slope deposit on the margins of such a basin (Okita, 1992). Another mode of manganese enrichment is suggested by the even greater ratio of manganese to iron in oceanic sediments and by the very high ratio in seawater. Both manganese and iron are leached from oceanic crust at mid-ocean ridge

294

Manganiferous Sediments, Rocks, and Ores commercial manganese concentrations (Gutzmer and Beukes 1996a, 1997). Therefore, we have two mechanisms for the addition of manganese to seawater: remobilization from reducing sediments and hydrothermal leaching of ridge-crest basalts. An interesting question emerges from consideration of these two processes: Is there any way to distinguish their products? Furthermore, has the ratio of the processes changed with time? 7.11.3 COMPOSITION OF MANGANESE ACCUMULATIONS

Figure 5 Distribution of dissolved Mn with depth in the Black Sea (reproduced by permission of Society of Economic Geologists from Sedimentary and Diagenetic Mineral Deposits: A Basin Analysis Approach to Exploration (eds. E. R. Force, J. J. Eidel, and J. B. Maynard), 1991, pp. 147– 159).

spreading centers and released to the overlying seawater. The iron reacts very rapidly with oxygen to produce iron oxides that fall out close to the vents. Manganese oxide formation is much slower, however, and manganese is dispersed much farther from the vent. Manganese anomalies in seawater are one way of tracking vent plumes which can travel hundreds of kilometers. This process generates an Mn/Fe vector in the sediments that can be used as a prospecting tool to find mineralized vents, as beautifully illustrated for the modern Red Sea by Ba¨cker et al. (1991). It is interesting to speculate that this process also accounts for the enrichment of Archean crust in manganese. Higher heat flow in the Archean would have resulted in a much higher flux from the ridge-crest hydrothermal systems, and a lowoxygen atmosphere would have prevented the accumulation of manganese oxides in deep-sea sediments, while still retaining much of the iron there as sulfides. Therefore, shallow marine deposits in the Archean should have been strongly enriched in manganese relative to iron. Veizer (1988, figure 5.23) has shown that manganese is strongly enriched in Archean carbonates, with a narrow peak at ,5,000 ppm, compared to Phanerozoic carbonates that show much lower values in a broad range from 10 ppm to 700 ppm. Manganese is so enriched in some older carbonates that their weathering produces lateritic residues with

Significant accumulations of manganese, which are defined here as containing .20% manganese, are widely distributed both in time and space. Representative analyses of some of the more prominent occurrences are presented in Table 2. Here a selection of large ore deposits and accumulations from various periods in the Earth’s history shows the general chemical character of these accumulations and how they differ with time and with environment. The most obvious differences are between modern deposits of the ocean floor that are remote from hydrothermal influences, the so-called hydrogenous deposits, and all of the other occurrences. The modern hydrothermal deposits are quite similar in many aspects to the ancient ore deposits, leading one to suspect a large hydrothermal component in the source of the manganese in the ancient deposits. The most prominent geochemical difference, and one that has been exploited in the classification of manganese deposits (see, e.g., Nicholson (1992)), is the extreme concentration of the heavy metals such as cobalt, nickel, and copper in the hydrogenous deposits. A comparison of the analyses in Table 2 for hydrogenous deposits to those for ancient deposits plus modern hydrothermal deposits shows a 10-fold or higher enrichment in the hydrogenous deposits for cobalt, nickel, and copper, but also for lead, thorium, and total rare earth elements (REEs). The ancient deposits and the modern hydrothermal deposits are similar for most elements; the ancient deposits show some enrichment in sulfur, arsenic, and selenium, whereas the modern hydrothermal deposits are relatively enriched in nickel, copper, and molybdenum. The extreme enrichment of metals in the hydrogenous deposits is usually attributed to their very slow accumulation from oxidizing seawater (Hein et al., 1997) compared to rapid deposition of manganese and iron from suddenly cooled hydrothermal fluids, again under oxidizing conditions. The rate of growth of hydrogenous crusts ranges from 0.5 mm Ma21 to 15 mm Ma21 compared to 20 mm Ma21 to 100 mm Ma21 for hydrothermal crusts. A consideration of the

Composition of Manganese Accumulations

295

Table 2 Composition of Mn ores and related accumulations. Deposit Constituent Mn Fe Na Mg Al Si P K Ca Ti C organic C carbonate d 13Corg d 13Corg d 18Ocarb S total d 34S Sc V Cr Co Ni Cu Zn As Se Rb Sr Y Zr Nb Mo Sn Sb Ba Pb Bi Th U La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Units

% 36.5 % 4.22 % 0.01 % 1.93 % 0.09 % 2.40 % 0.01 % 0.00 % 12.2 % 0.01 % % 3.15 per mil per mil 29.1 per mil 210.4 % 0.04 per mil ppm ppm 6 ppm 16 ppm 50 ppm 19 ppm ,5 ppm 74 ppm ppm ppm 1 ppm 146 ppm 5.3 ppm 2 ppm ,0.1 ppm ,2 ppm ,1 ppm ppm 360 ppm 6 ppm ppm 0.4 ppm 0.2 ppm 3.18 ppm 3.3 ppm 0.5 ppm 2.1 ppm 0.33 ppm 0.16 ppm 0.58 ppm 0.09 ppm 0.55 ppm 0.14 ppm 0.42 ppm 0.06 ppm 0.45 ppm 0.08

Sources a

Kalahari Tanganshan Xiangtan Molango Hydrothermal Hydrogenous Hydrogenous S. Africa China China Mexico Pacific Pacific Nodules Paleoprot. Neoprot. Neoprot. Jurassic Modern Modern Modern

Liu et al. (in press).

44.9 2.66 0.03 1.09 0.74 1.54 0.12 0.01 2.72 0.14 2.90 8.70

2.00 26.7 13 80 23 64 45 16 19 48 2.0 1 69 41 33 25.0 16.7 6.5 14 50 68 0.5 3.3 0.9 32.0 98.8 8.0 32.8 7.60 2.35 7.52 1.24 7.99 1.55 4.54 0.58 3.50 0.43

a,b b

Gutzmer (1996).

42.3 1.54 0.01 1.23 0.44 4.75 0.07 0.17 1.34 0.10 0.60 8.70

0.80 52.5 12 53 24 44 24 11 266 34 0.5 14 101 33 29 17.5 6.5 4.0 2.8 259 8 ,0.1 1.2 0.7 26 79.4 6.7 27.8 5.50 1.14 5.56 0.88 5.85 1.23 3.28 0.46 2.75 0.38

a c

Okita (1987).

28.0 8.53 0.00 5.10 1.43 5.36 0.06 0.00 1.73 0.06

Liu (1988).

22.1 15.1 1.6 1.26 1.01 3.69 1.18 0.56 4.13 0.77

18.6 12.5 1.7 1.6 2.7 7.7 0.25 0.7 2.3 0.67 0.1

2 27.3 2 13.1 2 3.6 0.20 3.2 2.12 67 15 132 60 7 48 31 ,0.2 40 8.6 15 1.5 2 ,1 1.1 45 6 1.4 1.6 11.8 18.1 2.4 9.5 1.70 0.43 1.87 0.29 1.49 0.31 0.85 0.11 0.75 0.15

a d

37.0 1.87 2.37 1.95 1.57 7.73 0.13 1.01 2.48 0.15

a,c,d e

0.06

0.305

3.8 225 48 72 287 228 238 33 0.12 ,0.2 555 17 23 9.9 327

6.4 515 22 6,400 5,400 1,080 680 165 0.4 ,0.2 1,210 166 172 ,0.1 445

25 1,380 45

24 1,700 1,780

0.7 2.1 18.9 16.3

33.0 9.6 202 1,100 106 162 42 9.90 26 7.53 57.8 6.60 31.9 4.30 17.7 3.34

7.2 0.99 0.28 0.25

0.78 0.14 e

Usui and Someya (1997).

e f

Li (2000).

0.47 10 500 35 2,700 6,600 4,500 1,200 140 0.6 17 830 150 560 50 400 2 40 2,300 900 7.0 30.0 5.0 157 530 36 158 35 9.00 32 5.40 31.0 7.00 18.0 2.30 20.0 1.80 f

296

Manganiferous Sediments, Rocks, and Ores

distribution of the REEs gives some insight into the relative contribution of hydrothermal fluids and seawater. Figure 6 shows that the hydrogenous accumulations have much higher REE concentrations and a pronounced positive cerium anomaly. The cerium anomaly is thought to result from the oxidation of Ce3þ to Ce4þ under the oxidizing conditions found in the modern ocean. Tetravalent cerium is much more strongly bound to Fe – Mn oxide surfaces than the trivalent form; cerium is therefore preferentially removed from seawater in the modern ocean; this leaves seawater with a pronounced negative cerium anomaly (Fleet et al., 1983). Oceanic hydrothermal fluids, mostly seawater, have an REE signature close to seawater and so produce deposits with a strong negative cerium anomaly, as seen in Figure 6. For the ancient deposits, both the Molango (Jurassic) and the Kalahari (Paleoproterozoic) deposits have REE plots very close to the hydrothermal signature. The Neoproterozoic deposits (Liu et al., in press), however, have flat REE distributions with a small positive cerium anomaly. This distribution could be obtained by a mixture of hydrothermal and hydrogenous sources. The model curves of Fleet et al. (1983, figure 9) suggest that 20 – 30% hydrogenous manganese mixed with a normal hydrothermal source would produce the cerium anomalies observed in the Neoproterozoic deposits, whereas the Molango and Kalahari deposits received less than 10% hydrogenous manganese. Europium anomalies have been used to infer a decline with time in the relative contributions of hydrothermal and hydrogenous sources in iron deposits. Many authors (e.g., Derry and Jacobsen, 1990; Bau and Mo¨ller, 1993; Klein

and Beukes, 1993) have argued that a strong positive anomaly indicates a hydrothermal source, and noted that the anomaly in iron formations declines sharply from the Archean to the Paleoproterozoic to the Neoproterozoic. By contrast, manganese deposits seem to be quite variable in their europium anomalies and to show no trend with time. For example, notice the mixed positive and negative europium anomalies for the Neoproterozoic deposits in Figure 6 and the strong positive anomaly for the young Molango deposit. In addition, modern hydrogenous nodules show a strong positive europium anomaly, inconsistent with their small hydrothermal component. This contrast between the behavior of iron and manganese oxide deposits is an area that needs more study. The enrichment of the ancient manganese deposits in sulfur-related elements such as arsenic and selenium compared to both the modern hydrothermal and hydrogenous crusts suggests that the ancient deposits formed under loweroxygen conditions, either in the bottom water of basins or in sediments. Low-oxygen bottom waters can be formed in restricted basins like the present Black Sea or could be caused by an overall low-oxygen ocean, perhaps from total freezing of a “snowball” Earth (Kirschvink et al., 2000). The strong negative cerium anomaly in the Kalahari samples, however, argues against a long period of whole-ocean anoxia, because no manganese nodules and crusts would have been present in deep water to sequester excess Ce4þ. Instead, large anoxic basins with restricted access to the open ocean seem to be the most likely source of manganese for large deposits (Okita, 1992; Okita and Shanks, 1992).

Figure 6 REE distribution for major Mn ore deposits shows variable Eu and Ce anomalies (source Table 2).

Common Manganese Minerals 7.11.4 COMMON MANGANESE MINERALS The mineralogy of manganese is complex and analytically difficult. Not only are a large number of phases commonly found, but they are difficult to characterize because of poor crystallinity, fine grain size, intimate intergrowths, and a propensity for alteration during sample handling. Oxides are the most common ore minerals. R. G. Burns and M. Burns (1977a,b) have suggested that manganese oxides consist of subunits of MnO82 6 octahedra. These are arranged in chains or sheets, much as in silicate structures, to give the observed mineral structures. In many cases the chains are arranged in such a way as to form large “tunnels” like those in zeolites. These spaces may be occupied by large cations such as Kþ, Ca2þ, or Ba2þ (Table 3). Complete descriptions of manganese phases, along with X-ray spacings and photomicrographs of polished sections, can be found in Frenzel (1980). See also Turner and Buseck (1983) for a useful summary of oxide structures as seen in transmission electron microscopy. The mineralogy of the Kalahari manganese field of South Africa has been studied thoroughly. The field is a good type area for examining the behavior of manganese minerals (see especially Kleyenstu¨ber (1986)). Gutzmer and Beukes (1996b) reported 135 mineral phases from the Kalahari deposits, 59 of which are manganese bearing. Of these 59, the oxides (bixbyite, braunite, hausmannite, jacobsite, and manganite) Table 3 Name Tetravalent oxides pyrolusite Ramsdellite Nsutite Hollandite Cryptomelane Psilomelane Todorokite Birnessite Lithiophorite Trivalent oxides Bixbyite Spinel structures Hausmannite Jacobsite Hydroxides Manganite Pyrochroite Silicates Braunite Carbonates Kutnahorite Rhodochrosite Source: Maynard (1983).

297

and the carbonates (kutnahorite and rhodochrosite) dominate. The original deposit contained braunite, hematite, and kutnahorite as a finely laminated sediment. The dominant early manganese oxide, braunite, comes in two varieties: type I with the usual structure Mn2þ Mn3þ 6 SiO12, and type II, which has less silicon and calcium and divalent manganese to make a structure Ca0.5(Mn3þ, Fe3þ)7Si0.5O12. Much of the original kutnahorite was converted to low-manganese calcite and hausmannite during regional metamorphism. Locally superimposed on these assemblages is a set of fault-related hydrothermal alterations. This hydrothermal activity produced a strong upgrading of the ore in total manganese by removal of CO2. The final product consists largely of hausmannite and hematite. The type II braunite appears to be a later replacement of the type I structure produced during this metasomatic alteration (Kleyenstu¨ber, 1986). The summary by Nel et al. (1986) gives an average manganese content of the unaltered ore as 37%, upgraded to 47% in the hydrothermally altered material, but with a reduction in Mn/Fe ratio from 13.3 to 3.8, suggesting considerable loss of manganese (or addition of iron) during alteration. Many deposits, including those of the Kalahari, contain manganese carbonates, and most other land-based deposits are thought to have originally had a large proportion of the manganese in carbonates. These phases are usually described as rhodochrosite, but microprobe analyses show the presence of considerable calcium.

Common Mn minerals.

Formula

Comments

b-MnO2 MnO2 d-MnO2 (Ba,K)1-2Mn8O16·x H2O K1 – 2Mn8O16·x H2O (Ba,K,Mn,Co)3(O,OH)6Mn8O16 (Na,Ca,K,Ba,Mn,)Mn3O7·x H2O (Ca,Na)(Mn2þMn4þ)7O14·x H2O (Al,Li)(OH)2MnO2

Single chains of (MnO6)28 octahedra Double chains Intergrowths of single and double Open tunnels permit large cation incorporation

a-(Mn,Fe)2O3

.30% Fe2O3

Mn3O4 Mn3O4

Up to 7 mol.% Fe3O4 .45 mol.% Fe3O4

g-MnOOH Mn(OH)2

Brucite structure

Mn7SiO12 CaMn(CO3)2 Mn CO3

¼ romanechite ˚ manganite ¼ 10 A ˚ manganite ¼7A Common in supergene zone

298

Manganiferous Sediments, Rocks, and Ores

Experimental work (Goldsmith and Graf, 1957) has revealed that there is a solubility gap in the Mn – Ca carbonates between 50 mol.% Mn (the mineral kutnahorite) and 80 mol.% Mn (calcian rhodochrosite), but many analyses of manganese carbonates from modern sediments show manganese contents that fall within this gap, suggesting that these phases are metastable with respect to kutnahorite and rhodochrosite. Changes in mineralogy during metamorphism have been described from the Kalahari manganese field of South Africa. Bixbyite, manganite, hausmannite, and a silica-deficient braunite are characteristic of the metamorphosed zone. Heubner (1976) suggested that the persistence of such minerals indicates the presence of locally high oxygen fugacities during metamorphism; manganosite is the only stable phase under conditions prevailing during the metamorphism of common ferruginous rocks. The Kalahari deposits were later subjected to supergene alteration, which resulted in the formation of a distinctive assemblage, including nsutite, todorokite, and lithiophorite. Like other transition elements, manganese is subject to crystal field effects, but only Mn4þ, and to a lesser extent the rare Mn3þ, have a crystal field stabilization energy. Accordingly, Mn4þ in octahedral positions in minerals should be strongly favored over Mn2þ in solution (Crerar et al., 1980, p. 296). Minerals that do contain Mn2þ, such as rhodochrosite, tend to be light colored, compared with the dark Mn4þ minerals, again because of the splitting of d orbitals of Mn4þ in the imposed crystal field, which gives rise to excited states with the same spin multiplicity.

7.11.5 BEHAVIOR OF MANGANESE IN IGNEOUS SETTINGS, ESPECIALLY MID-OCEAN RIDGE VENTS Igneous processes by themselves do not produce manganese enrichments because there is no discrimination between manganese and iron. More strongly oxidizing conditions are needed. A common place to find such conditions is where magmas reach the surface via volcanic activity. The high heat flow around volcanic centers induces hydrothermal circulation of the water in the surrounding rocks. Porphyry Cu –Mo deposits are a product of such circulation in terrestrial settings, where meteoric water is the circulating fluid, whereas Cyprus- or Kuroko-type deposits form in submarine settings with seawater as the fluid (Sillitoe, 1980). The seawater hydrothermal systems can precipitate sulfide minerals either within the volcanic rocks, within overlying sediments, or sometimes even at the surface where the fluids vent into a brine pool. In many

cases, however, the superficial sulfides are lost and the only permanent reminder of the presence of seafloor vents is an accumulation of Mn – Fe oxides. Beginning in the mid-1960s, geologists became aware of the deposition of metal-rich sediments along mid-ocean ridges (e.g., Bonatti and Joensuu, 1966; Bostro¨m and Peterson, 1966; see also Mills and Elderfield, 1995, figure 1). The prediction that the high heat flow in these areas should induce hydrothermal circulation of seawater through the oceanic crust (Elder, 1965; Deffeyes, 1970) made it logical to assume that these deposits were related to hydrothermal alteration of basalt by seawater. These predictions have been dramatically confirmed by the direct sampling and photography of active hydrothermal vents on a number of mid-ocean ridge vent sites. The volume of seawater flowing through such systems is large, and may exercise an important control on seawater chemistry (Maynard, 1976; Wolery and Sleep, 1988; Kadko et al., 1995). The dominant reactions seem to be exchange of Mg2þ in seawater for Ca2þ, Fe2þ, Mn2þ, and Ba2þ in the basalt. Considerable H4SiO4 is also released, and SO22 4 is precipitated as a constituent of CaSO4 and is reduced to H2S. In addition to these ridge crest, or axial systems, which operate at , 300 –400 8C, there are also off-axis systems that are cooler, ,100 – 150 8C. Manganese plumes seem to be confined to the high-temperature vents on the ridge crests (Murton et al., 1999). The insolubility of iron, manganese, and barium phases in seawater, under oxidizing conditions, leads to their precipitation and consequent enrichment in ridge-crest sediment. Because Fe2þ oxidizes at a lower Eh than Mn2þ, iron precipitates first in an Eh gradient, and manganese is dispersed farther from the vents. Thus, there is a facies sequence of Fe ! Mn that can be useful in exploration (e.g., Russell, 1975), and is one indicator of a seafloor– hydrothermal source for an orebody. Observations of modern seafloor vents suggest that plume dimensions can be on the order of hundreds of kilometers (e.g., Lupton, 1995) and that high manganese values can cover areas with the diameter on the order of 50 km (Murton et al., 1999, figure 9). The rapid precipitation of iron oxides close to the vents has a strong effect on the behavior of REE. The freshly precipitated iron oxides exhibit a very strong absorption of the REE, so that, despite the fact that vent fluids have perhaps 10 times the REE content of seawater, these are all removed close to the vents and the hydrothermal emission actually produces a net removal of REE from seawater (Mitra et al., 1994). Thus, at least in today’s high-O2 oceans, the REE signature of the seawater-hydrothermal systems is not transferred to the bulk seawater. The hydrothermal signature

Behavior of Manganese in Igneous Settings, Especially Mid-ocean Ridge Vents is characterized by a strong positive europium anomaly, produced by europium released from plagioclase during hydrothermal alteration. Modern iron deposits lack a positive europium anomaly unless they are deposited very close to the ridge crest (Olivarez and Owen, 1991), but manganese deposits show a wide variety of positive and negative anomalies (Figure 6; see also Graf et al., 1994; Fan et al., 1999). This variability indicates a strong local control of basin configuration on the proportion of hydrothermal and hydrogenous manganese. Experimental work on basalt –seawater reactions sheds further light on how hydrothermal Fe – Mn deposits form. For the mobilization of heavy metals from basalt, two factors seem to be important: temperature and the water/rock ratio (e.g., Seyfried and Bischoff, 1981; Seyfried and Ding, 1995). Either high temperatures (. 400 8C) or high water/rock ratios (.10 at 300 8C) are required. At the higher temperatures, complexing of the metals by the chloride in seawater becomes significant, and a large proportion of the metals in the basalt can be leached. At lower temperatures, however, the chloride content of seawater is too small for significant complexing, and no leaching takes place unless large amounts of Mg2þ pass through the rock. Because most of these systems are thought to operate at 300 – 400 8C (e.g., Hannington et al., 1995, figure 10), chloride complexing is probably not the important mechanism in leaching and transporting the metals; rather, Mg2þ reactions under high water/rock ratios dominate. There are several ancient analogues of midocean ridge deposits, the best known being the ophiolite-associated copper ore bodies of Cyprus (e.g., Robertson, 1975). Ophiolites, which are interpreted to be former oceanic ridges now exposed on land, have a characteristic stratigraphy. Metal enrichment occurs as epigenetic and syngenetic pyrite – chalcopyrite at the top of the sheeted dike complex and in the basal pillow lavas, and as syngenetic iron-rich or manganeserich accumulations either directly overlying the sulfides or higher in the pillow-lava sequence. The oxide portions of the Cyprus deposits occur in two forms: iron-rich deposits (ochres) directly overlying the massive sulfides and containing some intermixed sulfides, and manganese-rich deposits (umbers) higher in the section, within or at the top of the upper pillow lavas. The ochres are thought to be products of submarine alteration of the massive sulfides, and are similar in many respects to the hydrothermal oxide deposits of the East Pacific Rise. The umbers, in contrast, are richer in manganese, which imparts a chocolate-brown color, and are interbedded with pelagic sediment. Chemically, they are similar to the ochres, but with a higher Mn/Fe ratio. Because the umbers are

299

found higher in the stratigraphic sequence, separated from the ores by the upper pillow lavas, they are probably products of a separate event, but one related genetically to ore deposition. Robertson (1975, p. 528) proposed that they formed from hydrothermal solutions released into oxidizing seawater on the elevated flanks of the ridge, whereas the massive sulfides formed from solutions released into small anoxic basins within the axial rift. Subsequently, these became superficially oxidized to ochres. Fluid inclusions, strontium isotopes, and sulfur isotopes indicate that seawater was the hydrothermal fluid. Spooner and Bray (1977) showed that fluid inclusions in quartz co-precipitated with the ore have salinities indistinguishable from that of seawater. 87Sr/86Sr ratios for the Cyprus deposits, which in some samples are close to that of Cretaceous seawater, also indicate a large seawater component in the ore-forming fluids with water/rock ratios exceeding 15/1 (Chapman and Spooner, 1977; Spooner et al., 1977). Sulfur isotopes indicate that at least some sulfur from seawater was incorporated in the sulfides. As seawater SO22 4 , which starts at þ21 per mil in the hydrothermal recharge, is reduced to sulfide by the iron in basalt at high temperatures, there is an isotope fractionation that depends on the temperature and the extent of reaction (Ohmoto and Goldhaber, 1997). For example, reduction of 20% of the incurrent SO22 4 at 350 8C produces a final sulfide product with d 34S of þ 5 per mil, and a 40% extent of reaction results in þ 7 per mil. Modern deposits from ridges with minimal sediment cover have d 34S ranging from þ 1 per mil to þ 7 per mil compared to near 0 per mil for basalt-derived sulfur (Shanks et al., 1995). This range of values suggests that significant seawater and basalt-derived sulfur are both involved in the process. However, because the proportion of the incurrent SO22 4 that is reduced is unknown, the relative amounts cannot be quantified. The numbers are similar in ancient deposits. For the ophiolitic copper deposits of Notre Dame Bay, Newfoundland, Bachinski (1977) reported an average d 34S sulfide value of þ 9.0 per mil and for the Cyprus deposits of þ 4.8 per mil. Perhaps the best-studied deposits associated with oceanic ridges are those in the Red Sea, yet they are somewhat anomalous because of the proximity of continental landmasses, and the apparent involvement of continental material in their genesis. Good descriptions of the deposits and their inferred origin can be found in Shanks and Bischoff (1980) and Scholten et al. (2000). Overlying the metal-rich sediments is a dense, hot brine, from which they are believed to have been deposited. This lower brine is, in turn, separated from seawater by a second brine of intermediate composition, which is probably a mixture of the

300

Manganiferous Sediments, Rocks, and Ores

bottom brine and seawater. Chemical analyses of the brines show the depletion in Mg2þ and enrichment in Ca2þ, compared with seawater, that is typical of basalt – seawater interactions, but Naþ and Cl2 are unusually high. This enrichment is believed to be caused by dissolution of Miocene evaporites bordering the Red Sea. Of particular note is the greater dispersion of manganese than iron into the upper brine. Mapping of the Fe/Mn ratio in bottom sediments of the Atlantis II deep (Scholten et al., 2000, figure 14.5) shows that iron is highly concentrated near the vents, whereas manganese is much more widely dispersed. Furthermore, the Fe/Mn ratio provides a vector that could be used in locating vent positions in older deposits (Cronan, 1976). By assuming that the lower brine in the Atlantis II deep is a cooled sample of the hydrothermal fluid, Shanks and Bischoff (1977) have reconstructed the conditions of metal transport and deposition in this system. The results indicate that the brines were derived from normal seawater that acquired a high salinity through reaction with evaporites at low temperatures, then was heated to , 200 8C by interaction with recent intrusive rocks of the rift zone at high water/rock ratios. Heavy metals occur mostly as chloride complexes at higher temperatures, but cooling below 150 8C leads to dissociation and precipitation of the metals except manganese as sulfides. The high density of the brine layer restricts circulation, so that the bottom water becomes anoxic and the sulfides are protected from oxidation. An excess of metals over reduced sulfur should lead to precipitation of virtually all of the sulfur but considerable export of metals into surrounding sediments and seawater. Metalliferous sediments also occur around the volcanic islands of Stromboli and Santorini, as described by Bonatti et al. (1972), Puchelt (1973), and Puchelt et al. (1973). At Stromboli, the metalliferous sediment is mostly Fe– Mn oxide. X-ray diffraction shows only birnessite, but the presence of cryptocrystalline goethite and amorphous SiO2 is inferred. On Santorini, the metalliferous sediment is up to 3 m thick, has a high water content, and is X-ray amorphous. Little manganese is present, and CO2 is much more abundant than at Stromboli, indicating the presence of FeCO3. Sulfides are conspicuously rare at these two localities, but are common in sediments bordering the island of Vulcano, near Stromboli (Honnorez et al., 1973). Abundant fumaroles and hot springs occur on the island and in the surrounding water; the submarine activity is confined to depths of less than 15 m in an elongate zone , 100 m wide, parallel to the shoreline. Chemically, the fumarolic gases are mostly CO2 and H2O with , 0.04% H2S (Honnorez et al., 1973, table 1). Temperatures range from 100 8C to

600 8C, and mixing of the fumarolic gases with seawater lowers its pH to as low as 2. In the sediment, the result is a sulfidation of Fe – Ti grains, cementation of quartz sand grains by pyrite-marcasite, and silicification of volcanic rock fragments. No base-metal sulfides have been reported. Manganese is slightly enriched in the surrounding sediments, and reaches its highest amount, 1,900 ppm, in the deepest, most distal part of the bay in which the fumaroles occur (Valette, 1973, figure 3), a distribution consistent with the manganese dispersion seen in other deposits. Barium, however, does not follow manganese, but is highest (1,000 ppm) near the shoreline around the fumaroles (Valette, 1973, figure 7). 7.11.6 BEHAVIOR OF MANGANESE IN SEDIMENTATION Numerous examples are known of manganese enrichment in modern sediments that can be examined for analogues of the processes involved in the formation of manganese deposits in ancient rocks. It is convenient to treat these in two groups: (i) oxides, which form the deep-sea nodules and (ii) crusts, and carbonates, which are not as extensively developed in the modern, but are of interest because of the abundance of carbonate or carbonate-derived ores in the ancient. 7.11.6.1 Manganese Nodules and Crusts in Modern Sediments The world’s largest deposit of manganese is found in modern deep-sea sediments. Pelagic ferromanganese nodules cover the ocean floor over large areas, particularly in the central Pacific, and many volcanic edifices are coated by Fe – Mn crusts built on the rock surfaces. The nodules and crusts form the hydrogenetic end-member in what is really a continuous distribution of Fe – Mn accumulations between hydrogenetic and hydrothermal sources (Hein et al., 1997; Hein and Manheim, 2000). The nodules, in turn, can be divided into those that are supplied with metals from the overlying seawater and those that receive a contribution from diagenetic remobilization of metals from the sediment below. There is also a continuum for these processes and they may be significant for the same nodule (Halbach et al., 1982; Cronan, 1997). Mineralogically, the purely hydrogenetic crusts and nodules are dominated by g-MnO2 and amorphous FeOOH, epitaxially intergrown. Diagenetic nodules, alternatively, ˚ manganite (“todorokite”) and 7 A ˚ contain 10 A manganite (“birnessite”). The distinction between these two phases may be procedural only; in some ˚ phase can be collapsed to the 7 A ˚ cases the 10 A

Behavior of Manganese in Sedimentation phase by drying (Hein et al., 1997). The manganese and iron components of the crusts and nodules each contribute in different ways to the overall chemical properties of the bulk accumulation. For example, it is found that crusts from the central Pacific show a strong positive correlation between P REE and %Fe, but a negative correlation with %Mn. Ce/Ce*, alternatively, has a positive correlation with %Mn, but negative with %Fe. The nodules show interesting areal variations, particularly in minor-element chemistry (Calvert and Price, 1977; Cronan, 1997). Biologic productivity in the overlying surface waters exerts a strong influence on nodule compositions. Areas with low productivity, 40– 10 8N and 20 – 50 8S, have lower nickel and copper in the nodules, but higher iron and cobalt. The increase in Mn – Ni –Cu with increasing productivity is related to increased organic carbon in the sediments, which leads, in turn, to increased mobility for these elements under suboxic diagenetic conditions. Iron does not experience enhanced mobility because of the development of insoluble iron sulfides that tie up the iron within the sediment. Halbach et al. (1982) found that, on a microscale, cobalt is associated with silicate- and iron-rich microlaminae. They suggested that in the high electric field at the surface of Si – Fe colloids, Co2þ is oxidized to Co3þ, a form that is much more strongly bound in the iron-rich phase but is not able to substitute for manganese in the manganese-rich phase of the nodules. Waters with very high surface productivity such as are found near the equator, tend to show a reversal, and to have fewer nodules and less trace element enrichment. This decline may simply be dilution or it may be that conditions in the sediment and immediately adjacent bottom water are too reducing and the manganese and trace elements escape entirely. Manganese crusts act as a closed system with respect to the rare earths and a number of radioisotopes, which makes them a good recorder of changes in ocean chemistry (e.g., DeCarlo, 1991). Hein and Manheim (2000, figure 9.10) show how changes in neodymium, lead, and beryllium isotopes in manganese crusts track the circulation of deep water from its generation in the North Atlantic through the Indian Ocean and into the North Pacific.

301

7.11.6.2 Manganese Carbonates in Modern Sediments Because of the preponderance of manganese carbonates as ore minerals or as protores for deposits in ancient rocks, consideration needs to be given to modern-day accumulations of carbonates as well as oxides. Such occurrences have been described from deep-water sediments of the Panama Basin (Pedersen and Price, 1982), from shallow, near-shore sediments of Scotland (Calvert and Price, 1970), and from several areas of the Baltic (Suess, 1979; Glasby et al., 1997; Lepland and Stevens, 1998). In the pore waters of the sediments in the Panama Basin, concentrations of Mnþ2 reach values as high as 160 mM, and the dry sediments contain as much as 3 wt.% Mn. The high manganese concentrations are associated with manganese oxides at the sediment surface, but in one location a zone of manganese carbonate was also found at a depth of ,150 cm in the sediment. Whitish crusts at this depth proved to be coalesced microspheres of manganese carbonate, ,100 mm in diameter, with compositions close to those expected for kutnahorite (Table 4). No other diagenetic phases were found. Carbon isotopes (Table 4) suggest a negligible contribution of organic carbon to the manganese carbonates. The isotopic composition of carbon in the Panama Basin manganese carbonates is close to that of seawater and is probably derived entirely from pore-water HCO2 3 or the dissolution of shell material. In the Loch Fyne sediments, manganese occurs both in nodular masses of manganese oxide and as concretions of MnCO3, 1 –8 cm in diameter at water depths of 180 – 200 m. The oxide-rich nodules are commonly cemented and replaced by MnCO3 (Calvert and Price, 1970, figures 4 and 5). Over an area of ,10 km2, manganese concentrations in the surface sediment exceed 5%, sometimes reaching 10%, although the thickness of this surficial layer is only ,20 cm. Other constituents in the sediment are detrital quartz and clay and shell material. There do not appear to be distinct oxide and carbonate facies. The Loch Fyne sediments have carbon in MnCO3 that is somewhat lighter, isotopically, than carbon from the Panama Basin, indicating some contribution from decaying organic matter.

Table 4 Chemical and carbon isotopic composition of modern Mn carbonates. Locality Panama Basin Loch Fyne Baltic Sea Source: Maynard (1983).

Mn (mol.%)

Ca (mol.%)

Mg (mol.%)

d13C, per mil PDB

48 48 85

47 45 10

5 7 5

þ 2.6 2 5.8 2 13

302

Manganiferous Sediments, Rocks, and Ores

The Baltic sediments contain a complex diagenetic assemblage that includes among the reduced phases siderite, MnS, and iron phosphates, in addition to the manganese carbonates. Dissolved Mn2þ concentrations in the pore waters reach as much as 80 mM. Manganese carbonate occurs as microspheres 5 –25 mm across (Suess, 1979, figure 3), embedded in a matrix of amorphous silica. The MnS phase has an unusual hexagonal form, as seen with the scanning electron microscope. Bottcher and Huckriede (1997) reported that this phase has a sulfur isotopic composition of 2 13 per mil, indicating derivation from bacterial sulfate reduction. Cubic MnS (alabandite) is found in lesser amounts (Lepland and Stevens, 1998). The unusual presence of these manganese sulfides indicates that the environment was iron limited for pyrite formation, i.e., sulfur remained after all reactive iron was consumed (Sternbeck and Sohlenius, 1997). The carbon isotopic values from the MnCO3 (Table 4) are very negative and indicate that more than half of the carbon is organic derived. The dominance of organicderived carbon in these accumulations indicates that the MnCO3 formed diagenetically within the sediment. Oxides of manganese are also present in some isolated areas. Glasby et al. (1997) reported that abundant concretions of Mn – Fe oxides are found in three main areas of the Baltic: the Gulf of Bothnia, the Gulf of Finland, and the Gulf of Riga, with Mn/Fe ratios highest in the Gulf of Bothnia concretions. The Baltic is brackish in salinity. There is a strong contrast between the salinity of surface waters at 6.5 –7.5 ppt compared to that of inflowing bottom water from the North Sea at 15– 20 ppt. This density contrast leads to density stratification and the formation of a halocline that, in turn, promotes lowered oxygen in bottom waters and manganese mobility (Neumann et al., 2002). The three areas of Mn – Fe oxide accumulation are remote from the influence of the more saline bottom waters from the North Sea and tend to lack the stratification of the Baltic proper. The shallowness of the basins and the cold climate favor seasonal overturn of the water column. The higher bottom water oxygenation favors oxide development in these areas compared to the Baltic proper. Localized deeps within the main part of the Baltic tend to develop stable anoxic conditions, a situation that leads to a buildup of dissolved Mn2þ in the bottom water. Around the margins of these deeps, there is a zone of enrichment of manganese in the sediment near the halocline, where oxidation in the water column leads to deposition of manganese oxides that then convert to MnCO3 during burial. Oscillation of oxidation state, related to oscillations in the intensity of bottom-water inflows, can lead to laminated sediments with manganese-rich and

manganese-poor layers that provide a record of oxygenation events in the basin (Huckriede and Meischner, 1996; Burke and Kemp, 2002).

7.11.7 A GENERAL MODEL OF SEDIMENTARY MANGANESE MINERALIZATION The behavior of manganese in restricted marine basins such as the Black Sea and the Baltic suggests a general model of manganese behavior in which manganese is solubilized in deep-water sediments in anoxic basins and is reprecipitated around the margins of these basins at the point where the redox interface impinges on the seafloor. This model was first articulated by Force and Cannon (1988) from their observations of the modern and detailed facies analysis of a number of ancient deposits. Subsequently the model has been developed in some detail based on stable isotopic studies of Phanerozoic deposits, particularly Molango in Mexico (Okita, 1987, 1992; Okita et al., 1988; Maynard et al., 1990; Okita and Shanks, 1992). See also reviews by Force and Maynard (1991), who emphasized the ancient record and favored a dominant role for basin geometry, and by Calvert and Pedersen (1996), who emphasized the modern and argued for a dominant role of surface-water productivity in controlling manganese distribution. The process begins with the precipitation of manganese oxides within the water column at the interface between oxidizing and reducing conditions, usually a halocline. Most of the precipitated manganese simply redissolves as it passes downwards through the water column, unless the seafloor is shallow enough to intercept the redox interface (Figure 7). This phenomenon produces what might be called a “manganese compensation depth.” Below this depth, all particulate manganese is dissolved in the water column, so that none reaches the bottom and the sediments are low in manganese. At the depth of the halocline, there is a strong enrichment of the surface sediment in manganese oxide particles. At shallower depths, the sediments are again low in manganese, because the oxygenated surface portion of the water column contains virtually no dissolved manganese. Thus, there is a critical depth for manganese enrichment that produces a “bathtub ring” effect around the margins of the basin. Reaction with organic matter in the sediment then converts this manganese oxide to manganese carbonate. Iron is excluded from this cycling, because it is insoluble in the deeper-water sediments as the sulfide. A key observation supporting this model is a strong correlation between high manganese contents in the rocks and strongly negative carbon

Behavior of Manganese in Soils and Weathering

303

Figure 7 General model for Mn mineralization in euxinic basins.

isotopes (Okita and Shanks, 1992). The production of the MnCO3 mineralization, therefore, required the consumption of large amounts of organic matter and must have occurred during early diagenesis. The process can be represented schematically by the reaction 2MnO2 þ CH2 O þ HCO2 3 ! 2MnCO3 þ H2 O þ OH2

ð5Þ

From Table 2, the organic matter in the sediments has a d 13C of about 2 27.3. Combining equal amounts of this carbon with 0 per mil HCO2 3 from seawater would produce a carbonate mineral with d 13C of about 2 14 per mil, very close to the observed value. At the same time that the manganese is oxidizing the organic matter, it also attacks any iron sulfide in the sediment (Aller and Rude, 1988; Schippers and Jørgensen, 2001): FeS þ 4:5MnO2 þ 4H2 O ! FeOOH 2 þ 4:5 Mn2þ þ SO22 4 þ 7OH

ð6Þ

From this reaction, the deposit should be very low in sulfur, as is observed, and the pyrite that does form should be very heavy isotopically. This prediction of heavy sulfur is based on the requirement that any pyrite that forms be relatively late, forming after all of the manganese oxide has reacted. Therefore, the degree of contact with the overlying seawater reservoir of sulfate S will be limited, sulfate reduction will go to completion, and the small amount of sulfide that does form will be isotopically close to its parent sulfate compared to normal pyrite in black shale. Okita and Shanks (1992) reported d 34S values of pyrite from unmineralized black shales of 2 21 per mil compared to þ 3.2 per mil in the ore zones. Another prediction from this equation is that iron oxides should accompany the MnCO3 mineralization. The ore at Molango is, in fact, distinctly magnetic, and Okita (1992) has reported high concentrations of magnetite and maghemite. Subsequent work has shown that this model has broad applicability to manganese ore deposits.

See, for example, Nyame (1998) on the Nsuta deposit of Ghana, Tang and Liu (1999) for a representative Neoproterozoic deposit of China, and Tsikos (1999) for the Hotazel deposit in the Kalahari manganese field of South Africa. In addition, many deposits that had been thought to be dominated by primary manganese oxide may, in fact, be lateritic weathering products of manganese carbonate protores (Varentsov, 1996).

7.11.8 BEHAVIOR OF MANGANESE IN SOILS AND WEATHERING Manganese is an important nutrient in soils and can be a limiting factor in plant nutrition. Therefore, it has received considerable attention in the soil science community. Most of this literature focuses on the distribution of available rather than total manganese, which would be of more interest to geochemists trying to reconstruct past soil behavior. An additional limitation is that, because manganese is usually treated as a major element by geochemists but occurs in concentrations less than 1%, most soil analyses of manganese report only one or two significant figures, making judgments about manganese behavior in the soil difficult. The available literature for modern soils (see Maynard, 1992 and references therein) indicates that both iron and manganese have some mobility within the soil but that iron is always retained somewhere within the profile, while manganese may or may not be retained. Table 5 shows that iron and manganese occur in soils at a somewhat higher level than the corresponding parent rocks, but that the Mn/Fe ratio is relatively constant for all parent rock types in the range of 0.013 –0.016, similar to the range for crustal reservoirs and rock types given in Table 1. Sequential extraction of soils shows that the proportion of available to total manganese varies greatly, but averages ,50% (Alloway, 1995, p. 231). As predicted by the Eh – pH diagram for manganese, both lower pH and

0.016 0.020 0.019 0.016 0.026 0.025 0.025 0.027 0.013 58,500 86,200 10,400 13,800 84,900 32,900 93,100 29,200 23,800 930 1,700 200 220 2,200 820 2,300 800 320 0.016 0.015 0.013 0.009 0.023 0.007 0.020 0.014 0.011 101,000 119,000 13,600 14,300 97,800 116,000 101,000 37,900 25,900 1,600 1,800 180 130 2,300 840 2,000 520 290 Ultramafic Basalt Granite Granite Basalt Greenstone Basalt Ultramafic Granite Mt. Prinzera, Italy Belbex, France San Pedro, Portugal Llano, Texas Sturgeon, Canada Flinflon, Canada Quirke, Canada Kalkkloof, South Africa Ottosdol, South Africa

Modern Modern Modern 0.5 Ga 1.1 Ga 1.8 Ga 2.5 Ga 2.6 Ga 2.8 Ga

Mn/Fe host rock Fe (ppm) host rock Mn (ppm) host rock Mn/Fe soil Fe (ppm) soil Mn (ppm) soil Age of soil Host rock Name

Table 5 Distribution of Mn and Fe in soils and paleosols.

Venturelli et al. (1997) Chesworth et al. (1981) Middleburg et al. (1988) Capo (1984) Zbinden et al. (1988) Holland et al. (1989) Prasad and Roscoe (1991) Martini (1994) Grandstaff et al. (1986)

Manganiferous Sediments, Rocks, and Ores

Source

304

lower Eh favor manganese availability to plants. For example, manganese uptake by roots falls off sharply at pH values in the rhizosphere above 5.5. However, manganese availability rises sharply in the summer, presumably reflecting lowered O2 in the soil atmosphere. These strong shifts in manganese solubility with time and place in the soil provide the potential for considerable manganese migration. Most profiles show some removal of manganese from the surface layers of the soil. In some cases reprecipitation deeper retains all of this manganese within the profile, while other modern soils show considerable leaching. Middleburg et al. (1988) presented a geochemically detailed study of 10 soil profiles on the Iberian peninsula: five show manganese conserved and five show appreciable loss, whereas iron is conserved in all of the profiles. The average manganese loss over the 10 profiles is ,18%. Thus, modern soils tend to show some leaching of manganese with iron retained, reflecting the greater solubility of manganese in the presence of oxygen. To what extent is this true of ancient soils? Many paleosols have been examined geochemically to try to reconstruct atmospheric conditions of the past, particularly oxygen levels (see Maynard (1992) and Rye and Holland (1998) for reviews). In general, older paleosols, greater than ,2.2 Ga, show considerable iron depletion, whereas younger paleosols show conservation of iron within the profile, which has been interpreted to show that the atmosphere prior to 2.2 Ga held relatively little oxygen (Holland et al., 1989). Manganese, however, tends to show leaching in paleosols of all ages, as can be seen from the Mn/ Fe ratios for soils and host rocks given in Table 5, so that manganese behavior in soils is more uniform in time than that of iron. Manganese ore bodies generally have experienced supergene alteration that has produced a new set of minerals, and, via the removal of carbonate, produced higher grades of ore (Varentsov, 1996). Parc et al. (1989) provided a detailed assessment of the progress of weathering in manganese-rich rocks and proposed two sequences of increased oxidation states based on the dominant manganese mineral in the parent rock: rhodochrosite (2) ! manganite (3) ! cryptomelane (3.94) ! nsutite (3.95) ! lithiophorite (4) and Mn silicates (2) ! birnessite (3.71) ! nsutite (3.95) ! pyrolusite (4) where the number given in parentheses is the net oxidation state of manganese in the mineral. Cryptomelane is reported as a common phase in most weathering profiles on manganese ores and raises an interesting question of the source of the necessary potassium. Some can come from conversion of potassium silicates to kaolinite in the weathering profile, but even the

References potassium-free rocks of the Mamatwan ores are converted to a thick layer of cryptomelane (Gutzmer and Beukes, 1996b), which would seem to require some external source of potassium. A curious aspect of the weathering of manganese ore bodies is the occasional formation of “battery-active” MnO2, which is a far more valuable product than the ordinary metallurgical grade oxide. Usually the battery-active material is dominated by the mineral nsutite, but much nsutite does not exhibit this property. In fact, manganese exploration crews evaluate the grade of a deposit by making batteries in the field. Why some oxides exhibit this property and others do not has remained something of a mystery. Turner and Buseck (1983) have suggested that it is related to the presence of domains of triple-chain structure in some nsutites.

7.11.9 CONCLUSIONS Manganese is of great interest in geochemistry, because its minerals are both tracers of redox processes and accumulators of other elements of great geochemical significance. The solubility of manganese compared to iron under reducing and mildly oxidizing conditions leads to its export from low-oxygen environments, be they basalt – hydrothermal systems or euxinic sedimentary basins, and its accumulation in oxidizing environments of the shallow ocean or in low-productivity areas of the deep sea. Therefore, tracking Mn/Fe ratios provides us with a means of reconstructing the oxidation structure of ocean basins or of soils or of groundwater systems. The strong affinity of manganese oxides for certain transition elements and the REEs provides another way to gain insight into geochemical processes in the oceans. Especially valuable are the records of cerium and europium anomalies preserved in manganese accumulations, which can define the relative proportions of hydrothermal and diagenetic processes in the sedimentary record.

REFERENCES Aller R. C. and Rude P. D. (1988) Complete oxidation of solid phase sulfide by manganese and bacteria in anoxic marine sediments. Geochim. Cosmochim. Acta 52, 751– 765. Alloway B. J. (1995) Heavy Metals in Soils, 2nd edn. Blackie Academic and Professional Publishers, London. Bachinski D. J. (1977) Sulfur isotopic composition of ophiolitic cupriferous iron sulfide deposits, Notre Dame Bay, Newfoundland. Econ. Geol. 72, 243–257. Ba¨cker H., Marchig V., von Stackelberg U., Stoffers P., Puteanus D., and Tufar W. (1991) Hydrothermale Aktivita¨t

305

auf dem Meeresboden. Geologische Jahrbuch D93, 103 –197. Balzer W. (1982) On the distribution of iron and manganese at the sediment/water interface: thermodynamic versus kinetic control. Geochim. Cosmochim. Acta 46, 1153–1161. Bau M. and Mo¨ller P. (1993) Rare earth element systematics of the chemically precipitated component in Early Precambrian iron formations and the evolution of the terrestrial atmosphere – hydrosphere – lithosphere system. Geochim. Cosmochim. Acta 57, 2239–2249. Bonatti E. and Joensuu O. (1966) Deep-sea iron deposit from the South Pacific. Science 154, 643 –645. Bonatti E., Honnorez J., Joensuu O., and Rydell H. (1972) Submarine iron deposits from the Mediterranean Sea. In The Mediterranean Sea (ed. D. J. Stanley). Hutchinson and Ross, Stroudsburg, PA, pp. 701 –710. Bostro¨m K. and Peterson M. N. A. (1966) Precipitates from hydrothermal exhalations on the East Pacific Rise. Econ. Geol. 61, 1258–1265. Bottcher M. E. and Huckriede H. (1997) First occurrence and stable isotope composition of authigenic g-MnS in the central Gotland Deep (Baltic Sea). Mar. Geol. 137, 201 –205. Burke I. T. and Kemp A. E. S. (2002) Microfabric analysis of Mn-carbonate laminae deposition and Mn-sulfide formation in the Gotland Deep, Baltic Sea. Geochim. Cosmochim. Acta 66, 1589–1600. Burns R. G. and Brown B. (1972) Nucleation and mineralogical controls on the composition of manganese nodules. In Ferromanganese Deposits on the Ocean Floor (ed. D. R. Horn). Lamont-Doherty Geol. Observatory, New York, pp. 51–61. Burns R. G. and Burns M. (1977a) The mineralogy and crystal chemistry of deep-sea manganese nodules, a polymetallic resource of the twenty-first century. Phil. Trans. Roy. Soc. London 286A, 283–301. Burns R. G. and Burns M. (1977b) Mineralogy. In Marine Manganese Deposits (ed. G. Glasby). Elsevier, Amsterdam. Calvert S. E. and Pedersen T. F. (1996) Sedimentary geochemistry of manganese: implications for the environment of formation of manganiferous black shales. Econ. Geol. 91, 36– 47. Calvert S. E. and Price N. B. (1970) Composition of manganese nodules and manganese carbonates from Loch Fyne, Scotland. Contrib. Mineral. Petrol. 29, 215 –233. Calvert S. E. and Price N. B. (1977) Geochemical variation in ferromanganese nodules and associated sediments from the Pacific Ocean. Mar. Chem. 5, 43 –74. Capo R. C. (1984) Petrology and geochemistry of a Cambrian paleosol developed on Precambrian granite, Llano Uplift, Texas. MA Thesis, University of Texas at Austin (unpublished). Chapelle F. H. (1993) Ground Water Microbiology and Geochemistry. Wiley, New York. Chapman H. J. and Spooner E. T. C. (1977) 87Sr enrichment of ophiolitic sulphide deposits in Cyprus confirms ore formation by circulating seawater. Earth Planet. Sci. Lett. 35, 71 –78. Chesworth W., Dejou J., and Larroque P. (1981) The weathering of basalt and relative mobilities of the major elements at Belbex, France. Geochim. Cosmochim. Acta 45, 1235–1243. Cowen J. P. and Bruland K. W. (1985) Mineral deposits associated with bacteria. Implications for Fe and Mn marine biogeochemistry. Deep-Sea Res. 32, 253– 272. Crerar D. A., Cormick R. K., and Barnes H. L. (1980a) Geochemistry of manganese, an overview. In Geology and Geochemistry of Manganese (eds. I. M. Varentsov and Gy. Grasselly). E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, vol. 1, pp. 293–334.

306

Manganiferous Sediments, Rocks, and Ores

Cronan D. S. (1976) Implications of metal dispersion from hydrothermal systems for mineral exploration on mid-ocean ridges and in island arcs. Nature 262, 567–569. Cronan D. S. (1997) Some controls on the geochemical variability of manganese nodules with particular reference to the tropical South Pacific. In Manganese Mineralization: Geochemistry and Mineralogy of Terrestrial and Marine Deposits, Special Publication Number 119 (eds. K. Nicholson, J. R. Hein, B. Bu¨hn, and S. Dasgupta). Geological Society of London, pp. 139–152. DeCarlo E. H. (1991) Paleoceanographic implications of rare earth elements variability within a Fe–Mn crust from the central Pacific Ocean. Mar. Geol. 98, 449–467. Deffeyes K. S. (1970) The axial valley: a steady-state feature of the terrain. In The Megatectonics of Continents and Oceans (eds. H. Johnson and B. L. Smith). Rutgers University Press, Rutgers, NJ, pp. 194–222. Derry L. A. and Jacobsen S. B. (1990) The chemical evolution of Precambrian seawater: evidence from REEs in banded iron formations. Geochim. Cosmochim. Acta 54, 2965–2977. Dollhopf M. E., Nealson K. H., Simon D. M., and Luther G. W. (2000) Kinetics of Fe(III) and Mn(IV) reduction by the Black Sea strain of Shewanella putrefaciens using in situ solid state voltammetric Au/Hg electrodes. Mar. Chem. 70, 171 –180. Elder J. W. (1965) Physical processes in geothermal areas. In Terrestrial Heat Flow, Monograph No. 8 (ed. W. H. K. Lee). American Geophysical Union, Washington, DC, pp. 221–229. Fan D.-L., Ye J., Yin L.-M., and Zhang R.-F. (1999) Microbial processes in the formation of the Sinian Gaoyan manganese carbonate ore, Sichuan Province, China. Ore Deposit Rev. 15, 79–93. Fleet A. J., Bostro¨m K., Laubier L., and Smith K. L. (1983) Hydrothermal and hydrogenous ferro-manganese deposits: do they form a continuum? The rare earth element evidence. In Hydrothermal Processes at Seafloor Spreading Centers (ed. P. A. Rona). Plenum, New York, pp. 535 –555. Force E. R. and Cannon W. F. (1988) A depositional model for shallow-marine manganese deposits around black-shale basins. Econ. Geol. 83, 83–117. Force E. R. and Maynard J. B. (1991) Manganese: syngenetic deposits on the margins of anoxic basins. In Sedimentary and Diagenetic Mineral Deposits: A Basin Analysis Approach to Exploration (eds. E. R. Force, J. J. Eidel, and J. B. Maynard). Society of Econ. Geologists, El Paso, TX, pp. 147– 159. Frenzel G. (1980) The manganese ore minerals. In Geology and Geochemistry of Manganese (eds. I. M. Varentsov and Gy. Grasselly). E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, vol. 1, pp. 25–158. Froelich P. N., Klinkhammer G. P., Bender M. L., Luedtke N. A., Heath G. R., Cullem D., Dauphin P., Hammond D., Hartman B., and Maynard V. (1979) Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. Cosmochim. Acta 43, 1075–1091. Glasby G. (1974) Mechanisms of incorporation of manganese and associated trace elements in marine manganese nodules. Oceanogr. Mar. Biol. Ann. Rev. 12, 11–40. Glasby G. P., Emelyanov E. M., Zhamoida A., Baturin G. N., Leipe T., Bahlo R., and Bonacker P. (1997) Environments of formation of ferromanganese concretions in the Baltic Sea: a critical review. In Manganese Mineralization: Geochemistry and Mineralogy of Terrestrial and Marine Deposits, Special Publication No. 119 (eds. K. Nicholson, J. R. Hein, B. Bu¨hn, and S. Dasgupta). Geological Society of London, London, pp. 29–42. Goldsmith J. R. and Graf D. L. (1957) The system CaO– MnO– CO2: solid solution and decomposition relations. Geochim. Cosmochim. Acta 11, 310– 334. Graf J. L., O’Connor E. A., and Van Leeuwen P. (1994) Rare earth element evidence of origin and depositional

environment of Late Proterozoic ironstone beds and manganese-oxide deposits, SW Brazil and SE Bolivia. J. South Am. Earth Sci. 7, 115–133. Grandstaff D. E., Edelman M. J., Foster R. W., Zbinden E., and Kimberley M. M. (1986) Chemistry and mineralogy of Precambrian paleosols at the base of the Dominion and Pongola Groups (Transvaal, South Africa). Precamb. Res. 32, 97 –131. Gutzmer J. (1996) Genesis and alteration of the Kalhari and Postmasburg manganese deposits, Griqualand West, South Africa. PhD Dissertation, Rand Afrikaans University (unpublished). Gutzmer J. and Beukes N. J. (1996a) Karst-hosted fresh-water Paleoproterozoic manganese deposits, Postmasburg, South Africa. Econ. Geol. 91, 1435–1454. Gutzmer J. and Beukes N. J. (1996b) Mineral paragenesis of the Kalahari manganese field, South Africa. Ore Geol. Rev. 11, 405 –428. Gutzmer J. and Beukes N. J. (1997) Mineralogy and mineral chemistry of oxide-facies manganese ores of the Postmasburg manganese field, South Africa. Min. Mag. 61, 213–231. Halbach P., Giovanoli R., and von Borstel D. (1982) Geochemical processes controlling the relationship between Co, Mn, and Fe in early diagenetic deep-sea nodules. Earth Planet. Sci. Lett. 60, 226 –236. Hannington M. D., Jonasson I. R., Herzig M., and Petersen S. (1995) Physical and chemical processes of seafloor mineralization at mid-ocean ridges. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, pp. 115–157. Hein J. R. and Manheim F. T. (2000) Cobalt-rich ferromanganese crusts in the Pacific. In Handbook of Marine Mineral Deposits (ed. D. S. Cronan). CRC Press, Boca Raton, FL, pp. 239 –279. Hein J. R., Koschinsky A., Halbach P., Manheim F. T., Bau M., Kang J.-K., and Lubick N. (1997) Iron and manganese oxide mineralization in the Pacific. In Manganese Mineralization: Geochemistry and Mineralogy of Terrestrial and Marine Deposits, Special Publication Number 119 (eds. K. Nicholson, J. R. Hein, B. Bu¨hn, and S. Dasgupta). Geological Society of London, pp. 123– 138. Hem J. D. (1972) Chemical factors that influence the availability of iron and manganese in aqueous systems. Geol. Soc. Am. Spec. Pap. 140, 17–24. Hem J. D. (1981) Rates of manganese oxidation in aqueous systems. Geochim. Cosmochim. Acta 45, 1369–1374. Heubner J. S. (1976) The manganese oxides—a bibliographic commentary. In Oxide Minerals (ed. D. Rumble). Mineralogical Society of America Short Course Notes, vol. 3, pp. SH1 –SH17. Holland H. D., Feakes C. R., and Zbinden E. A. (1989) The Flin Flon paleosol and the composition of the atmosphere 1.8 BYB. Am. J. Sci. 289, 362–389. Honnorez J., Honnorez-Guerstein B., Valette J., and Wauschkuhn A. (1973) Present day formation of an exhalative sulfide deposit at Vulcano (Tyrrhenian Sea): Part II. Active crystallization of fumarolic sulfides in the volcanic sediments of the Baia di Levante. In Ores in Sediments (eds. G. C. Amstutz and A. J. Bernard). Springer, Heidelberg, pp. 139–166. Huckriede H. and Meischner D. (1996) Origin and environment of manganese-rich sediments within black-shale basins. Geochim. Cosmochim. Acta 60, 1399–1413. Kadko D., Baross J., and Alt J. (1995) The magnitude and global implications of hydrothermal flux. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, DC, pp. 446 –466. Kirschvink J. L., Gaidos E. J., Bertani L. E., Beukes N. J., Gutzmer J., Maepa L. N., and Steinberger R. E. (2000) Paleoproterozoic snowball Earth: extreme climatic and

References geochemical global change and its biological consequences. Proc. Natl. Acad. Sci. 97, 1400–1405. Klein C. and Beukes N. J. (1993) Sedimentology and geochemistry of the glaciogenic Late Proterozoic Rapitan iron-formation in Canada. Econ. Geol. 88, 542–565. Kleyenstu¨ber A. S. E. (1986) The mineralogy of the manganese-bearing Hotazel Formation, of the proterozoic transvaal sequence in Griqualand West, South Africa. In Mineral Deposits of Southern Africa (eds. C. R. Anhaeusser and S. Maske). Geological Society of South Africa, Johannesburg, pp. 257–272. Krauskopf K. B. (1957) Separation of manganese from iron in sedimentary processes. Geochim. Cosmochim. Acta 12, 61–84. Langmuir D. (1997) Aqueous Environmental Chemistry. Prentice-Hall, Upper Saddle River, NJ. Lepland A. and Stevens R. L. (1998) Manganese authigenesis in the Landsort Deep, Baltic Sea. Mar. Geol. 151, 1–25. Li Y.-H. (2000) A Compendium of Geochemistry. Princeton University Press, Princeton, NJ. Liu T.-B. (1988) C – S – Fe correlation of shales hosting sedimentary manganese deposits. PhD Dissertation, University of Cincinnati (unpublished). Liu T.-B., Maynard J. B., and Alten J. Superheavy S isotopes from glacial-associated sediments of the Neoproterozoic of South China: oceanic Anoxia or sulfate limitation? Geol. Soc. Am. Spec. Pap. (in press). Lovley D. R. and Phillips E. J. P. (1988) Manganese inhibition of microbial iron reduction in anaerobic sediments. Geomicrobiol. J. 6, 145 –155. Lupton J. E. (1995) Hydrothermal plumes: near and far field. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, DC, pp. 317 –346. Mandernack K. W. and Tebo B. M. (1993) Manganese scavenging and oxidation at hydrothermal vents and in vent plumes. Geochim. Cosmochim. Acta 57, 3907–3923. Mandernack K. W., Post J., and Tebo B. M. (1995) Manganese mineral formation by bacterial spores of the marine Bacillus, strain SG-1: evidence for the direct oxidation of Mn(II) to Mn(IV). Geochim. Cosmochim. Acta 59, 4393– 4408. Martini J. E. J. (1994) A Late-Archean–Palaeoproterozoic (2.6 Ga) palaeosol on ultramafics in the Eastern Transvaal, South Africa. Precamb. Res. 67, 159 –180. Maynard J. B. (1976) The long-term buffering of the oceans. Geochim. Cosmochim. Acta 40, 1523–1532. Maynard J. B. (1983) Geochemistry of Sedimentary Ore Deposits. Springer, Heidelberg. Maynard J. B. (1992) Chemistry of modern soils as a guide to interpreting Precambrian paleosols. J. Geol. 100, 279 –289. Maynard J. B., Okita P. M., May E. D., and Martinez-Vera A. (1990) Palaeogeographic setting of Late Jurassic manganese mineralization in the Molango District, Mexico. In Sediment Hosted Mineral Deposits, International Association of Sedimentologists, Special Publication 11 (eds. J. Parnell, L.-J. Ye, and C.-M. Chen). Blackwell, Boston, pp. 17–30. Middleburg J. J., van der Weijden C. H., and Woitties J. R. W. (1988) Chemical processes affecting the mobility of major, minor and trace elements during weathering of granitic rocks. Chem. Geol. 68, 253–273. Mills R. A. and Elderfield H. (1995) Hydrothermal activity and the geochemistry of metalliferous sediment. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, DC, pp. 392 –407. Mitra A., Elderfield H., and Greaves M. J. (1994) Rare earth elements in submarine hydrothermal fluids and plumes form the Mid-Atlantic Ridge. Mar. Chem. 46, 217–235. Moffett J. W. and Ho J. (1996) Oxidation of cobalt and manganese in seawater via a common microbially

307

catalyzed pathway. Geochim. Cosmochim. Acta 60, 3415–3424. Murton B. J., Redbourn L. J., German C. R., and Baker E. T. (1999) Sources and fluxes of hydrothermal heat, chemicals and biology within a segment of the Mid-Atlantic Ridge. Earth Planet. Sci. Lett. 171, 301–317. Myers C. R. and Nealson K. H. (1988) Microbial reduction of manganese oxides: interactions with iron and sulfur. Geochim. Cosmochim. Acta 52, 2727–2732. Nealson K. H. (1997) Sediment bacteria: Who’s there, what are they doing, and what’s new? Ann. Rev. Earth Planet. Sci. 25, 403 –434. Nel D. J., Beukes N. J., and De Villiers J. P. R. (1986) The Mamatwan manganese mine of the Kalhari manganese field. In Mineral Deposits of Southern Africa (eds. C. R. Anhaeusser and S. Maske). Geological Society of South Africa, Johannesburg, pp. 963–978. Neumann T., Heiser U., Leosson M. A., and Kersten M. (2002) Early diagenetic processes during Mn carbonate formation: evidence from the isotopic composition of authigenic Carhodochrosite of the Baltic Sea. Geochim. Cosmochim. Acta 66, 867–879. Nicholson K. (1992) Contrasting mineralogical–geochemical signatures of manganese oxides: guides to metallogenesis. Econ. Geol. 87, 1253–1264. Nyame F. K. (1998) Mineralogy, geochemistry, and genesis of the Nsuta manganese deposit, Ghana. PhD Dissertation, Okayama University (unpublished). Ohmoto H. and Goldhaber M. B. (1997) Sulfur and carbon isotopes. In Geochemistry of Hydrothermal Ore Deposits (ed. H. L. Barnes). Wiley, New York, pp. 517– 611. Okita P. M. (1987) Geochemistry and mineralogy of the Molango manganese orebody, Hidalgo State, Mexico. PhD Dissertation, University of Cincinnati (unpublished). Okita P. M. (1992) Manganese carbonate mineralization in the Molango District, Mexico. Econ. Geol. 87, 1345–1366. Okita P. M. and Shanks W. C. (1992) Origin of stratiform sediment-hosted manganese carbonate ore deposits: examples from Molango, Mexico, and TaoJiang, China. Chem. Geol. 99, 139 –164. Okita P. M., Maynard J. B., Spiker E. C., and Force E. R. (1988) Isotopic evidence for organic matter oxidation by manganese reduction in the formation of stratiform manganese carbonate ore. Geochim. Cosmochim. Acta 52, 2679–2685. Olivarez A. M. and Owen R. M. (1991) The europium anomaly of seawater: implications for fluvial versus hydrothermal REE inputs to the oceans. Chem. Geol. 92, 317–328. Parc S., Nahon D., Tardy Y., and Vieillard P. (1989) Estimated solubility products and fields of stability for cryptomelane, nsutite, birnessite, and lithiophorite based on natural lateritic weathering sequences. Am. Mineral. 74, 466–475. Pedersen T. F. and Price N. B. (1982) The geochemistry of manganese carbonate in Panama Basin sediment. Geochim. Cosmochim. Acta 46, 59– 68. Prasad N. and Roscoe S. M. (1991) Profiles of altered zones at ca. 2.45 Ga unconformities beneath Huronian strata, Elliot Lake, Ontario: evidence for Early Aphebian weathering under anoxic conditions. Geol. Surv. Can., Pap. 91-1C, 43 –54. Puchelt H. (1973) Recent iron sediment formation at the Kameni Islands, Santorini (Greece). In Ores in Sediments (eds. G. C. Amstutz and A. J. Bernard). Springer, Heidelberg, pp. 227– 246. Puchelt H., Schock H. H., and Schroll E. (1973) Rezente marine Eisenerze auf Santorin, Greichenland: I. Geochemie, Entstehung, Mineralogie. Geol. Rundsch. 62, 786– 803. Robertson A. H. F. (1975) Cyprus umbers: basalt–sediment relationships on a Mesozoic ocean ridge. J. Geol. Soc. London. 131, 511–531. Roitz J. S., Flegal A. R., and Bruland K. W. (2002) The biogeochemical cycling of manganese in San Francisco Bay:

308

Manganiferous Sediments, Rocks, and Ores

temporal and spatial variations in surface water concentrations. Estuar. Coast. Shelf Sci. 54, 227–239. Roy S. (1981) Manganese Deposits. Academic Press. Russell M. J. (1975) Lithogeochemical environment of the Tynagh basemetal deposit, Ireland, and its bearing on ore deposition. Trans. Inst. Mining Metal. 84, B128–B133. Rye R. and Holland H. D. (1998) Paleosols and the evolution of atmospheric oxygen: a critical review. Am. J. Sci. 298, 621 –672. Schippers A. and Jørgensen B. B. (2001) Oxidation of pyrite and iron sulfide by manganese dioxide in marine sediments. Geochim. Cosmochim. Acta 65, 915– 922. Scholten J. C., Stoffers P., Garbe-Scho¨nberg D., and Moammar M. (2000) Hydrothermal mineralization in the Red Sea. In Handbook of Marine Mineral Deposits (ed. D. S. Cronan). CRC Press, Boca Raton, FL, pp. 369–395. Seyfried W. E. and Bischoff J. L. (1981) Experimental seawater –basalt interaction at 300 8C, 500 bars: chemical exchange, secondary mineral formation and implications for the transport of heavy metals. Geochim. Cosmochim. Acta 45, 135–147. Seyfried W. E. and Ding K. (1995) Phase equilibria in subseafloor hydrothermal systems: a review of the role of redox, temperature, pH and dissolved Cl on the chemistry of hot spring fluids at mid-ocean ridges. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, DC, pp. 248 –272. Shanks W. C. and Bischoff J. L. (1977) Ore transport and deposition in the Red Sea geothermal system: a geochemical model. Geochim. Cosmochim. Acta 41, 1507– 1519. Shanks W. C. and Bischoff J. L. (1980) Geochemistry, sulfur isotope composition, and accumulation rates of Red Sea geothermal deposits. Econ. Geol. 75, 445–459. Shanks W. C., Bo¨hlke J. K., and Seal R. R. (1995) Stable isotopes in mid-ocean ridge hydrothermal systems: interactions between fluids, minerals, and organisms. In Seafloor Hydrothermal Systems, Monograph 91 (eds. S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, and R. E. Thomson). American Geophysical Union, Washington, DC, pp. 194 –221. Sillitoe R. H. (1980) Are porphyry copper and kuroko-type massive sulfide deposits incompatible? Geology 8, 11–14. Sivaprakash C. (1980) Mineralogy of manganese deposits of Kodura and Garbham, Andhra Pradesh, India. Econ. Geol. 75, 1083–1104. Spooner E. T. C. and Bray C. J. (1977) Hydrothermal fluids of seawater salinity in ophiolitic sulphide ore deposits in Cyprus. Nature 266, 808–812. Spooner E. T. C., Chapman H. J., and Smewing J. D. (1977) Strontium isotope contamination and oxidation during ocean floor hydrothermal metamorphism of the ophiolitic rocks of the Troodos Massif, Cyprus. Geochim. Cosmochim. Acta 41, 873 –890. Sternbeck J. and Sohlenius G. (1997) Authigenic sulfide and carbonate mineral formation in Holocene sediments of the Baltic Sea. Chem. Geol. 135, 55– 73. Stumm W. and Morgan J. J. (1996) Aquatic Chemistry, 3rd edn. Wiley-Interscience, New York, 1022pp.

Suess E. (1979) Mineral phases formed in anoxic sediments by microbial decomposition of organic matter. Geochim. Cosmochim. Acta 43, 339–352. Tang S.-Y. and Liu T.-B. (1999) Origin of the early Sinian Minle manganese deposit, Hunan Province, China. Ore Geol. Rev. 15, 71–78. Tebo B. M. and Nealson K. H. (1984) Microbial mediation of Mn(II) and Co(II) precipitation at the O2/H2 S interface in two anoxic fjords. Limnol. Oceanogr. 29, 1247–1258. Tsikos H. (1999) Petrographic and geochemical constraints on the origin and post-depositional history of the Hotazel iron – manganese deposits, Kalahari Manganese Field, South Africa. PhD Dissertation, Rhodes University (unpublished). Turner S. and Buseck P. R. (1983) Defects in nsutite (g-MnO2) and dry-cell battery efficiency. Nature 304, 143–146. Usui A. and Someya M. (1997) Distribution and composition of marine hydrogenetic and hydrothermal manganese deposits in the northwest Pacific. In Manganese Mineralization: Geochemistry and Mineralogy of Terrestrial and Marine Deposits, Special Publication No. 119 (eds. K. Nicholson, J. R. Hein, B. Bu¨hn, and S. Dasgupta). Geological Society of London, London. pp. 177–198. Valette J. N. (1973) Distribution of certain trace elements in marine sediments surrounding Vulcano Island (Italy). In Ores in Sediments (eds. G. C. Amstutz and A. J. Bernard). Springer, Heidelberg, pp. 321–338. Varentsov I. M. (1996) Manganese Ores of Supergene Zone: Geochemistry of Formation. Kluwer Academic, Dordrecht. Veizer J. (1988) The evolving exogenic cycle. In Chemical Cycles in the History of the Earth (eds. C. B. Gregor, R. M. Garrels, F. T. Mackenzie, and J. B. Maynard). WileyInterscience, pp. 175–220. Venturelli G., Contini S., and Bonazzi A. (1997) Weathering of ultramafic rocks and element mobility at Mt. Prinzera, Northern Apennines, Italy. Min. Mag. 61, 765–778. Vrobelsky D. A. and Chapelle F. H. (1994) Temporal and spatial changes of terminal electron-accepting processes in a petroleum hydrocarbon-contaminated aquifer and the significance for contaminant biodegradation. Water Resour. Res. 30, 1561–1570. Wijsman J. W. M., Herman P. M. J., Middleburg J. J., and Soetaert K. (2002) A model for early diagenetic processes in sediments of the continental shelf of the Black Sea. Estuar. Continent. Shelf Sci. 54, 403–421. Wolery T. J. and Sleep N. D. (1988) Interactions of geochemical cycles with the mantle. In Chemical Cycles in the History of the Earth (eds. C. B. Gregor, R. M. Garrels, F. T. Mackenzie, and J. B. Maynard). Wiley-Interscience, New York, pp. 77–104. Zbinden E. A., Holland H. D., and Feakes C. R. (1988) The Sturgeon Falls paleosol and the composition of the atmosphere 1.1 Ga BP. Precamb. Res. 42, 141 –163. Zhang J., Lion L. W., Nelson Y. M., Shuler M. L., and Ghiorse W. C. (2002) Kinetics of Mn(II) oxidation by Leptothrix discophora SS1. Geochim. Cosmochim. Acta 65, 773–781.

q 2003, Elsevier Ltd. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher.

Treatise on Geochemistry ISBN (set): 0-08-043751-6 Volume 7; (ISBN: 0-08-044342-7); pp. 289– 308